scispace - formally typeset
Search or ask a question
Journal ArticleDOI

Fast robust dose calculation on GPU for high-precision 1 H, 4 He, 12 C and 16 O ion therapy: the FRoG platform

TL;DR: FRoG, Fast Recalculation on GPU, enables comparative analysis of different models for estimation of physical and biological effective dose in 3D within minutes and in excellent agreement with the gold standard Monte Carlo simulation.
Abstract: Radiotherapy with protons and heavier ions landmarks a novel era in the field of high-precision cancer therapy. To identify patients most benefiting from this technologically demanding therapy, fast assessment of comparative treatment plans utilizing different ion species is urgently needed. Moreover, to overcome uncertainties of actual in-vivo physical dose distribution and biological effects elicited by different radiation qualities, development of a reliable high-throughput algorithm is required. To this end, we engineered a unique graphics processing unit (GPU) based software architecture allowing rapid and robust dose calculation. FRoG, Fast Recalculation on GPU, currently operates with four particle beams available at Heidelberg Ion Beam Therapy center, i.e., raster-scanning proton (1H), helium (4He), carbon (12C) and oxygen ions (16O). FRoG enables comparative analysis of different models for estimation of physical and biological effective dose in 3D within minutes and in excellent agreement with the gold standard Monte Carlo (MC) simulation. This is a crucial step towards development of next-generation patient specific radiotherapy.

Content maybe subject to copyright    Report

Citations
More filters
Journal ArticleDOI
TL;DR: Here, it is demonstrated that by combing two (or more) ions in a single field arrangement, more robust biological and more conformal dose distributions can be delivered compared to conventional particle therapy treatment planning.
Abstract: Purpose To develop and validate combined ion-beam with constant relative biological effectiveness (RBE) (CICR) particle therapy in single field arrangements for improved treatment efficacy, robustness, and normal tissue sparing. Methods and Materials The PRECISE (PaRticle thErapy using single and Combined Ion optimization StratEgies) treatment planning system was developed to investigate clinical viability of CICR treatments. Single-field uniform dose (SFUD) with a single ion (proton [p], helium [He], or carbon [C]) and CICR (C-p and C-He) treatments were generated for 3 patient cases with a clinically prescribed dose of 3 Gy (RBE) per fraction. Spread-out Bragg peak plans were irradiated in homogenous and clinical-like settings using an anthropomorphic head phantom. A dosimetric and biological verification of CICRC-p treatments using a murine glioma cell line (GL261) was performed. Results CICR treatment plans for the 3 patients presented highly uniform physical dose while reducing high dose-averaged linear energy transfer gradients compared with carbon ions alone. When considering uncertainty in tissue parameter (α/β)x assignment and RBE modeling, the CICRC-p treatment exhibited enhanced biophysical stability within the target volume, similar to protons alone. CICR treatments reduced dose to normal tissue surrounding the target, exhibiting similar or improved dosimetric features compared with SFUDHe. For both CICRC-p and SFUD treatments, measurements verified the planned dose in the target within ∼3%. Planned versus measured target RBE values were 1.38 ± 0.02 and 1.39 ± 0.07 ( Conclusions Here, we demonstrate that by combining 2 (or more) ions in a single field arrangement, more robust biological and more conformal dose distributions can be delivered compared with conventional particle therapy treatment planning. This work constitutes the first dosimetric and biological verification of multi-ion particle therapy in homogeneous as well as heterogenous settings.

39 citations

Journal ArticleDOI
TL;DR: Existing phenomenological and mechanistic/biophysical models were successfully integrated and validated in both Monte Carlo and GPU-accelerated analytical platforms against in vitro experiments, and tested using pristine peaks and clinical fields in highly radio-resistant tissues where models exhibit the greatest RBE uncertainty.
Abstract: Helium (4He) ion beam therapy provides favorable biophysical characteristics compared to currently administered particle therapies, i.e., reduced lateral scattering and enhanced biological damage to deep-seated tumors like heavier ions, while simultaneously lessened particle fragmentation in distal healthy tissues as observed with lighter protons. Despite these biophysical advantages, raster-scanning 4He ion therapy remains poorly explored e.g., clinical translational is hampered by the lack of reliable and robust estimation of physical and radiobiological uncertainties. Therefore, prior to the upcoming 4He ion therapy program at the Heidelberg Ion-beam Therapy Center (HIT), we aimed to characterize the biophysical phenomena of 4He ion beams and various aspects of the associated models for clinical integration. Characterization of biological effect for 4He ion beams was performed in both homogenous and patient-like treatment scenarios using innovative models for estimation of relative biological effectiveness (RBE) in silico and their experimental validation using clonogenic cell survival as the gold-standard surrogate. Towards translation of RBE models in patients, the first GPU-based treatment planning system (non-commercial) for raster-scanning 4He ion beams was devised in-house (FRoG). Our data indicate clinically relevant uncertainty of ±5–10% across different model simulations, highlighting their distinct biological and computational methodologies. The in vitro surrogate for highly radio-resistant tissues presented large RBE variability and uncertainty within the clinical dose range. Existing phenomenological and mechanistic/biophysical models were successfully integrated and validated in both Monte Carlo and GPU-accelerated analytical platforms against in vitro experiments, and tested using pristine peaks and clinical fields in highly radio-resistant tissues where models exhibit the greatest RBE uncertainty. Together, these efforts mark an important step towards clinical translation of raster-scanning 4He ion beam therapy to the clinic.

38 citations

Journal ArticleDOI
TL;DR: RBE dependencies in vitro, in vivo and in silico were investigated with respect to various clinically relevant endpoints in the context of tumor-specific tissue radio-sensitivity assignment and accurate RBE modeling.
Abstract: Purpose Present-day treatment planning in carbon ion therapy is conducted with assumptions for a limited number of tissue types and models for effective dose. Here, we comprehensively assess relative biological effectiveness (RBE) in carbon ion therapy and associated models toward the modernization of current clinical practice in effective dose calculation. Methods Using 2 human (A549, H460) and 2 mouse (B16, Renca) tumor cell lines, clonogenic cell survival assay was performed for examination of changes in RBE along the full range of clinical-like spread-out Bragg peak (SOBP) fields. Prediction power of the local effect model (LEM1 and LEM4) and the modified microdosimetric kinetic model (mMKM) was assessed. Experimentation and analysis were carried out in the frame of a multidimensional end point study for clinically relevant ranges of physical dose (D), dose-averaged linear energy transfer (LETd), and base-line photon radio-sensitivity (α/β)x. Additionally, predictions were compared against previously reported RBE measurements in vivo and surveyed in patient cases. Results RBE model prediction performance varied among the investigated perspectives, with mMKM prediction exhibiting superior agreement with measurements both in vitro and in vivo across the 3 investigated end points. LEM1 and LEM4 performed their best in the highest LET conditions but yielded overestimations and underestimations in low/midrange LET conditions, respectively, as demonstrated by comparison with measurements. Additionally, the analysis of patient treatment plans revealed substantial variability across the investigated models (±20%-30% uncertainty), largely dependent on the selected model and absolute values for input tissue parameters αx and βx. Conclusion RBE dependencies in vitro, in vivo, and in silico were investigated with respect to various clinically relevant end points in the context of tumor-specific tissue radio-sensitivity assignment and accurate RBE modeling. Discovered model trends and performances advocate upgrading current treatment planning schemes in carbon ion therapy and call for verification via clinical outcome analysis with large patient cohorts.

33 citations

Journal ArticleDOI
23 Oct 2018-Cancers
TL;DR: FRoG patient-specific quality assurance (QA) has been performed and matched the measured QA dosimetric data within ~2.0% for both particle species, demonstrating FRoG’s potential to bolster clinical activity with proton and carbon ion beams at CNAO.
Abstract: A fast and accurate dose calculation engine for hadrontherapy is critical for both routine clinical and advanced research applications. FRoG is a graphics processing unit (GPU)-based forward calculation tool developed at CNAO (Centro Nazionale di Adroterapia Oncologica) and at HIT (Heidelberg Ion Beam Therapy Center) for fast and accurate calculation of both physical and biological dose. FRoG calculation engine adopts a triple Gaussian parameterization for the description of the lateral dose distribution. FRoG provides dose, dose-averaged linear energy transfer, and biological dose-maps, -profiles, and -volume-histograms. For the benchmark of the FRoG calculation engine, using the clinical settings available at CNAO, spread-out Bragg peaks (SOBPs) and patient cases for both proton and carbon ion beams have been calculated and compared against FLUKA Monte Carlo (MC) predictions. In addition, FRoG patient-specific quality assurance (QA) has been performed for twenty-five proton and carbon ion fields. As a result, for protons, biological dose values, using a relative biological effectiveness (RBE) of 1.1, agree on average with MC within ~1% for both SOBPs and patient plans. For carbon ions, RBE-weighted dose (DRBE) agreement against FLUKA is within ~2.5% for the studied SOBPs and patient plans. Both MKM (Microdosimetric Kinetic Model) and LEM (Local Effect Model) DRBE are implemented and tested in FRoG to support the NIRS (National Institute of Radiological Sciences)-based to LEM-based biological dose conversion. FRoG matched the measured QA dosimetric data within ~2.0% for both particle species. The typical calculation times for patients ranged from roughly 1 to 4 min for proton beams and 3 to 6 min for carbon ions on a NVIDIA® GeForce® GTX 1080 Ti. This works demonstrates FRoG’s potential to bolster clinical activity with proton and carbon ion beams at CNAO.

30 citations


Cites background or methods or result from "Fast robust dose calculation on GPU..."

  • ...The FRoG dose calculation engine adopts a triple Gaussian distribution model for handling lateral dose distribution as a function of the water equivalent depth [24]....

    [...]

  • ...In general, the worsening of the agreement in the case of carbon ions is due to the limitations of handling the lateral dose distributions with a simple triple Gaussian approach [24] as well as the assumption that, for biological calculations, lateral variations of the mixed radiation field are not explicitly considered [26]....

    [...]

  • ...6% for patient cases (see Supplementary Material Table S1), in line with results presented previously [24] where calculations of SOBPs in homogenous settings using FRoG were within 1....

    [...]

  • ...The improved agreement compared to Syngo is due to the employment of a triple Gaussian parameterization in FRoG instead of a double Gaussian model as in Syngo [24] and an improved handling of the RS for proton beams....

    [...]

  • ...As described previously [24], physical and biophysical quantities were simulated and scored depth-wise in water....

    [...]

Journal ArticleDOI
TL;DR: Fast plan generation with a clinically validated dose calculation and optimizer is a promising approach for daily adaptive proton therapy, as well as for automated or highly interactive planning.
Abstract: Background: Treatment planning for intensity modulated proton therapy (IMPT) can be significantly improved by reducing the time for plan calculation, facilitating efficient sampling of the large so...

25 citations

References
More filters
Journal ArticleDOI
TL;DR: ROOT, written in C++, contains an efficient hierarchical OO database, a C++ interpreter, advanced statistical analysis (multi-dimensional histogramming, fitting, minimization, cluster finding algorithms) and visualization tools.
Abstract: The ROOT system in an Object Oriented framework for large scale data analysis. ROOT written in C++, contains, among others, an efficient hierarchical OO database, a C++ interpreter, advanced statistical analysis (multi-dimensional histogramming, fitting, minimization, cluster finding algorithms) and visualization tools. The user interacts with ROOT via a graphical user interface, the command line or batch scripts. The command and scripting language is C++ (using the interpreter) and large scripts can be compiled and dynamically linked in. The OO database design has been optimized for parallel access (reading as well as writing) by multiple processes.

4,586 citations

Journal ArticleDOI
TL;DR: Title of program: MINUIT gating the shape of the function near the minimum to determine the errors, in this essentially statistical problem the Catalogue number: ACWH function involved is usually a chisquare or negative log-likeli

2,947 citations

DOI
01 Jan 2005
TL;DR: The 2005 version of the Fluka particle transport code is described in this article, where the basic notions, modular structure of the system, and an installation and beginner's guide are described.
Abstract: This report describes the 2005 version of the Fluka particle transport code. The first part introduces the basic notions, describes the modular structure of the system, and contains an installation and beginner's guide. The second part complements this initial information with details about the various components of Fluka and how to use them. It concludes with a detailed history and bibliography.

2,271 citations

Journal ArticleDOI
TL;DR: The FLUKA Monte Carlo code as discussed by the authors is used extensively at CERN for all beam-machine interactions, radioprotection calculations and facility design of forthcoming projects, which requires the code to be consistently reliable over the entire energy range (from MeV to TeV) for all projectiles.

1,511 citations

Journal ArticleDOI
TL;DR: The framework and guidance is provided to allow radiation oncology physicists to design comprehensive and practical treatment planning QA programs for their clinics, and the scope of the QA needs for treatment planning is quite broad, encompassing image-based definition of patient anatomy.
Abstract: In recent years, the sophistication and complexity of clinical treatment planning and treatment planning systems has increased significantly, particularly including three-dimensional (3D) treatment planning systems, and the use of conformal treatment planning and delivery techniques. This has led to the need for a comprehensive set of quality assurance (QA) guidelines that can be applied to clinical treatment planning. This document is the report of Task Group 53 of the Radiation Therapy Committee of the American Association of Physicists in Medicine. The purpose of this report is to guide and assist the clinical medical physicist in developing and implementing a comprehensive but viable program of quality assurance for modern radiotherapy treatment planning. The scope of the QA needs for treatment planning is quite broad, encompassing image-based definition of patient anatomy, 3D beam descriptions for complex beams including multileaf collimator apertures, 3D dose calculation algorithms, and complex plan evaluation tools including dose volume histograms. The Task Group recommends an organizational framework for the task of creating a QA program which is individualized to the needs of each institution and addresses the issues of acceptance testing, commissioning the planning system and planning process, routine quality assurance, and ongoing QA of the planning process. This report, while not prescribing specific QA tests, provides the framework and guidance to allow radiation oncology physicists to design comprehensive and practical treatment planning QA programs for their clinics.

812 citations